Implementation of a Framework for Imitation Learning on a Humanoid Robot Using a Cognitive Architecture

Robots are designed to assist human beings to complete tasks. As described in many research journals and scientific fictions, researchers and the general public expect that one day robots can complete certain tasks independently and autonomously in our world. A typical implementation of Artificial Intelligence (AI) in robotics research is expected to incorporate Knowledge Representation, Automated Reasoning, Machine Learning, and Computer Vision [Russell and Norvig, 2010]. From the beginning of this century, the importance of cognitive sciences is highlighted by the researchers in both robotics and artificial intelligences[Russell and Norvig, 2010]. Real Cognitive Science is based on the testing and experiments on animals and humans to obtain the understanding and knowledge of the cognition, and AI research is based on designing and testing algorithms on a computer based artificial system. In the development, research in both Cognitive Sciences and AI benefit each other. Cognitive Science provides theoretical foundation and solutions to the problems in AI, and AI enhances the research in Cognitive Science and provides possible research directions for Cognitive Science. Recently, an emerging field, named Cognitive Robotics, is developed for robots to generate human-like behaviors and intelligences, which integrates perception, action, learning, decision-making, and communication[Kawamura and Browne, 2009]. Cognitive Robotics largely incorporates the concepts in the research of cognitive sciences and tries to simulate the architecture of the information processing of human cognition. Currently, some long term requirements are proposed for Cognitive Robotics, which put concentration on the embodiment of cognitive concepts [Anderson, 2003; Sloman et al., 2006]. Increasing achievements in Cognitive Robotics began to grab the attention of the robotics research community. However, currently, it is still difficult to design truly cognition, especially human-like cognition, on a robotic platform due to the limitations in mechanism, computation, architecture, etc. [Brooks, 1991]. Therefore, on one side, researchers try to place robots in a human-existing dynamic environment to complete tasks[Brooks, 1991]; on the other side, researchers still do not obtain a general architecture to generate complex behaviors in such area for robots[Sloman, 2001]. In recent research, researchers try to place robots in robotic aid area, where robots can assist humans to complete tasks[Tan et al., 2005].

The Proof of Concept of a Shadow Robotic System for Independent Living at Home

In the framework of the EU funded SRS (Multi-Role Shadow Robotic System for independent Living) project, an innovative semi autonomous service robot is under development with the aim to support frail elderly people at their home. This paper reports about the user validation of the SRS concept involving 63 potential users of the system coming from Italy, Germany and Spain: in particular they were frail elderly people, their relatives and 24 hour telecare professionals. Results confirmed that monitoring and managing emergency situations as well as helping with reaching, fetching and carrying objects that are too heavy or positioned in unreachable places are the tasks for which a robot is better accepted to address users' needs. To support the scenarios executions and operation modes, the interaction concept should provide three different interaction devices and modalities for each user group.

Development of rescue robotic systems for both daily and emergency use

In order to improve our life further, it is our inspiration to develop efficient robotic systems that can assist our usual need. The system that we are developing not only bring forward efficiency and credibility for daily usage or research platform, but when natural catastrophes like earthquake happens to strike, our robotic system is also capable of entering post-disaster sites and collect information about victim states that will be transmitted to the human rescuers, so that a more accurate and efficient rescue mission can be carried out. This paper presents the up-to-date information about our robotic system development of two different types of robots, the snake-type and the wheel-type.

A Design Methodology for Abstracting Character Archetypes onto Robotic Systems

Given the increase in robotic systems for the household, it is imperative to design the expressive modes of these systems, that in turn engender likability, animacy, acceptance, and trust. This paper approaches this problem by proposing a design methodology that can be used to abstract archetypal characters across the system, including form factor, user instructions, and interactive modalities. This approach uses Laban Movement Analysis paired with the Kansei Engineering iterative design approach to dissect movement and visual traits of archetypal characters and marry them to features of the robot. These character traits are explored in a product, channel, consumer framework and are realized through interface elements, such as color, animated eyes, and character specific motion profiles. Finally, the use of priming using familiar characters from popular culture as a means to enhance the recognition of character traits is explored. Results show that users associated traits specific to each character archetype that were consistent with the intended design and that an aggregate measure of interaction success went up with character priming. This was bolstered in the priming cases, where users rated these traits more strongly. Thus, this methodology serves as a tool to create meaningful design variations to robotic systems using character archetypes, allowing us to design user-specific personality traits and interactive elements.

An Experimental Evaluation of Bayesian Soft Human Sensor Fusion in Robotic Systems

As we move forward into the twenty-rst century, we are seeing new horizons opening to us through the use of autonomous robots as explorers, going places we can’t or won’t. The interaction between the human operator and robotic agent is of paramount importance in exploring these new horizons. Both humans and robots have their strengths and weaknesses, and this paper explores how they compliment each other and looks at new, more ecient ways to facilitate communication between the robot and its operator. This paper investigates conditions to the use of Bayesian information fusion algorithms with Gaussian mixture models to fuse soft human input with robotic sensor data to complete cooperative mission objectives. The condition investigated here are the benets of training individual models for the human input, or if a generic model trained on many people is sucient to complete the mission; we are also seeking to determine if it is useful to allow the human operator to assign a condence value to the information sent to the robot. These two conditions are evaluated with sizable number of human subjects working with a robotic platform to complete a multi-target search mission. Along with the results of these two conditions, reactions to the limited interaction with a robotic partner of a variety of people will be presented based on a survey that was performed once the trials were completed.

Requirement-oriented Configuration of Parallel Robotic Systems

Facing the increasing cost pressure, the decreasing product development time and demanded flexibility in today’s markets various strategies are conceivable. Within the area of high dynamic parallel robots the use of flexible modular systems for configuration and reconfiguration is one solution approach. In combination with an improved handling of requirements (i.e. a suitable structuring of requirements for the different phases of design process and hierarchy of product elements) and the modeling of inter-model relations (e.g. between requirement, structure and cost models) customer-oriented products can be build up fast.

Robust Backstepping Control of Robotic Systems Using Neural Networks

Neural network (NN) controllers for the robust back stepping control of robotic systems in both continuous and discrete-time are presented. Control action is employed to achieve tracking performance for unknown nonlinear system. Tuning methods are derived for the NN based on delta rule. Novel weight tuning algorithms for the NN are obtained that are similar to ε-modification in the case of continuous-time adaptive control. Uniform ultimate boundedness of the tracking error and the weight estimates are presented without using the persistency of excitation (PE) condition. Certainty equivalence is not used and regression matrix is not computed. No learning phase is needed for the NN and initialization of the network weights is straightforward. Simulation results justify the theoretical conclusions.

Reliability-Based Design Optimization of Robotic System Dynamic Performance

In this investigation a robotic system's dynamic performance is optimized for high reliability under uncertainty. The dynamic capability equations (DCE) allow designers to predict the dynamic performance of a robotic system for a particular configuration and reference point on the end effector (i.e., point design). Here the DCE are used in conjunction with a reliability-based design optimization (RBDO) strategy in order to obtain designs with robust dynamic performance with respect to the end-effector reference point. In this work a unilevel performance measure approach is used to perform RBDO. This is important for the reliable design of robotic systems in which a solution to the DCE is required for each constraint call. The method is illustrated on a robot design problem.

Testing of the mechatronic robotic system of the differential lock control on a truck

This article deals with testing the algorithm developed for autonomous control lock of the differential in a truck. Hardware for hardware in the loop testing of robotic systems or automotive systems is used and the proposed control algorithm was implemented into it. The system was applied to a vehicle with 8 × 8 drive and control algorithm evaluates input sensor values to set robotic console control. It is an actuator which consists of electrovalve, pneumatic cylinder, console, special dog clutch and feedback limit switch. This robotic console is used to lock the truck differentials. Verification of the correct algorithm operation was performed by manoeuvres and the response of the whole system was monitored. The aim was to get the reaction times of this mechatronic robotic system and evaluate them.

The Development of a Robotic System for Maintenance and Inspection of Power Lines

This paper describes preliminary studies on a prototype robot for the inspection/maintenance of electrical power lines. Faulty insulators or abnormal objects can be detected before causing a failure in the distribution of electrical energy. The robot uses a statically stable variation of the brachistochrone motion to move along the power lines and overcome the standard obstacles present in the lines. The paper presents the kinematics and preliminary simulation results on the dynamics and control of the robot.

Building a swarm of robotic bees

Swarm Robotics refers to the application of Swarm Intelligence techniques where a desired collective behavior emerges from the local interactions of robots with one another and with their environment. In this paper, a modified Bees Algorithm is proposed for multi-target search and coverage by an autonomous swarm of robotic “bees”. The objective is to find targets in an unknown area, send their estimated locations and fitness values to other robots in swarm which then provide the coverage of the found targets in a self-organized, decentralized way. The robots are equipped with ultrasonic sensors for obstacle avoidance, thermal sensors for target detection, and ZigBee modules for local communication. For the experiments, a small swarm of robots was built to test the performance of the modified Bees Algorithm. The experimental results show that the swarm is self-organized, decentralized and adaptive, and it can be successfully applied to the unknown area search and coverage.

Evaluation of a CT-Guided Robotic System for Precise Percutaneous Needle Insertion.

PURPOSE To assess overall targeting accuracy for CT-guided needle insertion using prototype robotic system for common target sites. MATERIALS AND METHODS Using CT guidance, metallic (2 × 1 mm) targets were embedded in retroperitoneum (n = 8), kidneys (n = 8), and liver (n = 14) of 8 Yorkshire pigs (55-65 kg). Bronchial bifurcations were targeted in the lung (n = 13). CT datasets were obtained for planning and controlled needle placement of commercially available 17- to 19-gauge needles (length 15-20 cm) using a small, patient-mounted, CT-guided robotic system with 5° of motion. Mean distance to target was 92.9 mm ± 19.7 (range, 64-146 mm). Planning included selection of target, skin entry point, and 4.6 ± 1.3 predetermined checkpoints (range, 2-9) where additional CT imaging was performed to permit stepwise correction of needle trajectory path as needed. Scanning and needle advancement were coordinated with breath motion using respiratory gating. Accuracy was assessed as distance from needle tip to predefined target. RESULTS Of 45 needle insertions performed, 2 were unsuccessful owing to technical issues. Accuracy of targeting was 1.2-1.4 mm ± 0.6 for kidney, retroperitoneum, and lung (P = .51), with 2.9 mm ± 1.9 accuracy for liver (P = .0003). This was achieved in 39 cases (91%) using a single insertion. Intraprocedural target movement was detected (3.5 mm ± 2.1 in retroperitoneum and 6.4 mm ± 3.9 in liver); the system compensated for 52.9% ± 30.3 of this movement. One pneumothorax was the only complication (8%). CONCLUSIONS Accurate needle insertion (< 3 mm error) can be achieved in common target sites when using a CT-guided robotic system. Stepwise checks with corrective angulation can potentially overcome issues of target movement during a procedure from organ deformity and other causes.

Robotic technologies for fast deployment of industrial robot systems

The development of breakthrough technologies helps the deployment of robotic systems in the industry. The implementation and integration of such technologies will improve productivity, flexibility and competitiveness, in diverse industrial settings specially for small and medium enterprises. In this paper we present a framework that integrates three novel technologies, namely safe robot arms with multi-modal and auto-calibrated sensing skin, a robot control framework to generate dynamic behaviors fusing multiple sensor signals, and an intuitive and fast teaching by demonstration method that segments and recognizes the robot activities on-line based on re-usable semantic descriptions. In order to validate our framework, these technologies are integrated in a industrial setting to sort and pack fruits. We demonstrate that our presented framework enables a standard industrial robotic system to be flexible, modular and adaptable to different production requirements.

An efficient and generic 2D Inevitable Collision State-checker

An inevitable collision state (ICS) for a robotic system is a state for which, no matter what the future trajectory of the system is, a collision eventually occurs. ICS can be used for both motion planning (to reduce the search space) and reactive navigation (for obvious safety reasons, a robotic system should never ever move to an ICS). ICS are particularly suited for navigation in dynamic environments since they take into account the future behaviour of the moving objects. Using ICS in practice is difficult given the intrinsic complexity of their characterization. The main contribution of this paper is a generic and efficient ICS-Checker, ie an algorithm that determines whether a given state is an ICS or not, for planar robotic systems with arbitrary dynamics moving in dynamic environments. The efficiency is obtained by applying the following principles: (a) reasoning on 2D slices of the state space of the robotic system, (b) precomputing off-line as many things as possible, and (c) exploiting graphics hardware performances. The ICS-Checker has been applied to two different robotic systems: a car-like vehicle and a spaceship. It has also been integrated in a reactive navigation scheme to safely drive the car-like vehicle.

A Robotic System for Overlapping Radiofrequency Ablation in Large Tumor Treatment

Overlapping radiofrequency ablation (RFA) in large tumor treatment requires multiple insertions of electrodes, which often compromise its efficacy and predictability. Surgical robot is a promising candidate for the execution of multiple RFA in large tumor treatment in terms of accuracy and speed. In this paper, we share our experience of design and implementation of a novel robotic system specialized for overlapping RFA. It consists of two components: a robotic manipulator and an automatic ablation planning module. The manipulator architecture is designed to facilitate the kinematic requirement for the multiple overlapping ablation technique within the constraints of minimally invasive surgery. An efficient “Voxel Growing” algorithm is adopted to automatically produce the ablation points according to the tumor's profile. The feasibility of the proposed robotic system is demonstrated by extensive simulation- and experiment-based evaluation conducted on ex vivo porcine liver.

Hybrid robotic systems for upper limb rehabilitation after stroke: A review.

In recent years the combined use of functional electrical stimulation (FES) and robotic devices, called hybrid robotic rehabilitation systems, has emerged as a promising approach for rehabilitation of lower and upper limb motor functions. This paper presents a review of the state of the art of current hybrid robotic solutions for upper limb rehabilitation after stroke. For this aim, studies have been selected through a search using web databases: IEEE-Xplore, Scopus and PubMed. A total of 10 different hybrid robotic systems were identified, and they are presented in this paper. Selected systems are critically compared considering their technological components and aspects that form part of the hybrid robotic solution, the proposed control strategies that have been implemented, as well as the current technological challenges in this topic. Additionally, we will present and discuss the corresponding evidences on the effectiveness of these hybrid robotic therapies. The review also discusses the future trends in this field.

Fundamental Collective Behaviors in Swarm Robotics

Passivity-Based Control and Estimation in Networked Robotics

Highlighting the control of networked robotic systems, this book synthesizes a unified passivity-based approach to an emerging cross-disciplinary subject. Thanks to this unified approach, readers can access various state-of-the-art research fields by studying only the background foundations associated with passivity. In addition to the theoretical results and techniques,the authors provide experimental case studies on testbeds of robotic systems including networked haptic devices, visual robotic systems, robotic network systems and visual sensor network systems.The text begins with an introduction to passivity and passivity-based control together with the other foundations needed in this book. The main body of the book consists of three parts. The first examines how passivity can be utilized for bilateral teleoperation and demonstrates the inherent robustness of the passivity-based controller against communication delays. The second part emphasizes passivitys usefulness for visual feedback control and estimation. Convergence is rigorously proved even when other passive components are interconnected. The passivity approach is also differentiated from other methodologies. The third part presents the unified passivity-based control-design methodology for multi-agent systems. This scheme is shown to be either immediately applicable or easily extendable to the solution of various motion coordination problems including 3-D attitude/pose synchronization, flocking control and cooperative motion estimation.Academic researchers and practitioners working in systems and control and/or robotics will appreciate the potential of the elegant and novel approach to the control of networked robots presented here. The limited background required and the case-study work described also make the text appropriate for and, it is hoped, inspiring to students.

Adverse Events in Robotic Surgery: A Retrospective Study of 14 Years of FDA Data

Background Use of robotic systems for minimally invasive surgery has rapidly increased during the last decade. Understanding the causes of adverse events and their impact on patients in robot-assisted surgery will help improve systems and operational practices to avoid incidents in the future. Methods By developing an automated natural language processing tool, we performed a comprehensive analysis of the adverse events reported to the publicly available MAUDE database (maintained by the U.S. Food and Drug Administration) from 2000 to 2013. We determined the number of events reported per procedure and per surgical specialty, the most common types of device malfunctions and their impact on patients, and the potential causes for catastrophic events such as patient injuries and deaths. Results During the study period, 144 deaths (1.4% of the 10,624 reports), 1,391 patient injuries (13.1%), and 8,061 device malfunctions (75.9%) were reported. The numbers of injury and death events per procedure have stayed relatively constant (mean = 83.4, 95% confidence interval (CI), 74.2–92.7 per 100,000 procedures) over the years. Surgical specialties for which robots are extensively used, such as gynecology and urology, had lower numbers of injuries, deaths, and conversions per procedure than more complex surgeries, such as cardiothoracic and head and neck (106.3 vs. 232.9 per 100,000 procedures, Risk Ratio = 2.2, 95% CI, 1.9–2.6). Device and instrument malfunctions, such as falling of burnt/broken pieces of instruments into the patient (14.7%), electrical arcing of instruments (10.5%), unintended operation of instruments (8.6%), system errors (5%), and video/imaging problems (2.6%), constituted a major part of the reports. Device malfunctions impacted patients in terms of injuries or procedure interruptions. In 1,104 (10.4%) of all the events, the procedure was interrupted to restart the system (3.1%), to convert the procedure to non-robotic techniques (7.3%), or to reschedule it (2.5%). Conclusions Despite widespread adoption of robotic systems for minimally invasive surgery in the U.S., a non-negligible number of technical difficulties and complications are still being experienced during procedures. Adoption of advanced techniques in design and operation of robotic surgical systems and enhanced mechanisms for adverse event reporting may reduce these preventable incidents in the future.

Aggregation in Swarm Robotic Systems: Evolution and Probabilistic Control

In this study we investigate two approachees for aggregation behavior in swarm robotics systems: Evolutionary methods and probabilistic control. In first part, aggregation behavior is chosen as a case, where performance and scalability of aggregation behaviors of perceptron controllers that are evolved for a simulated swarm robotic system are systematically studied with different parameter settings. Using a cluster of computers to run simulations in parallel, four experiments are conducted varying some of the parameters. Rules of thumb are derived, which can be of guidance to the use of evolutionary methods to generate other swarm robotic behaviors as well. In the second part a systematic analysis of probabilistic aggregation strategies in swarm robotic systems is presented. A generic aggregation behavior is proposed as a combination of four basic behaviors: obstacle avoidance, approach, repel, and wait. The latter three basic behaviors are combined using a three-state finite state machine with two probabilistic transitions among them. Two different metrics were used to compare performance of strategies. Through systematic experiments, how the aggregation performance, as measured by these two metrics, change 1) with transition probabilities, 2) with number of simulation steps, and 3) with arena size, is studied. We then discuss these two approaches for the aggregation problem.